- cross-posted to:
- aiphotography@lemmings.world
- cross-posted to:
- aiphotography@lemmings.world
cross-posted from: https://no.lastname.nz/post/275253
“Streetpunk Sexy” - Visualize a portrait of Neo. Neo is a young 22 year old woman. Neo is on an medieval street in a small village where the (second stories jut over the street)+++. The streets are cobbled.it is rainy afternoon,The shot, taken from a distance, captures Neo walking away in the golden hour. Neo is wearing a long and flowing red riding cape with detailed leaf and vine patterns stitched into it over a (black gothic dress:2) .The humid atmosphere ads mist to the street and the (mountains in the distance):2. Long dark messy hair escapes the red hood of Neo’s cape, The image showcases the natural late afternoon light, in hyperrealistic detail under soft lighting. The depth of field focuses on her against a backdrop of soothing,muted tones and high contrast,creating a dark,yet gorgeous (1.2) and immersive evening scene.
Negative prompt: text, watermark, low-quality, signature, moiré pattern, downsampling, aliasing, distorted, blurry, glossy, blur, jpeg artifacts, compression artifacts, poorly drawn, low-resolution, bad, distortion, twisted, excessive, exaggerated pose, exaggerated limbs, grainy, symmetrical, duplicate, error, pattern, beginner, pixelated, fake, hyper, glitch, overexposed, high-contrast, bad-contrast
Steps: 26, Sampler: DPM++ 2M SDE Karras, CFG scale: 7, Seed: 995203940, Face restoration: CodeFormer, Size: 1024x512, Model hash: a19b862e79, Model: fullyREALXL_v30ForREAL, Pad conds: True, Version: v1.7.0
![]()
This is more working on the prompt from https://no.lastname.nz/post/273813
Why do you specify “young” and “22 year old”?
SD will add more weight to terms that are repeated, or rephrased. (EDIT: or at least that is in at least one source that I’ve read)
young woman:
22 year old woman:
Both terms:
Seems like random output to me. Try one term three times in a row and see if you get similar variance. Content-wise I see no age indicators in the silhouette.
As I mentioned in the other post, the prompt is based on the one from the image below: https://civitai.com/images/4771726. I’ll have a tinker with the start of the prompt and see what it does - we are all here to learn